I was recently involved in a discussion at work about the definition of a certain word. Some folks were resistant to my attempt to define this term. And there were several comments along the lines of “I know it when I see it.”
I respectfully disagree. You don’t know it when you see it. I truly believe that concept is an epistemological absurdity. In fact, I think the reverse – “I see it when I know it” – more faithfully represents both human knowledge and human vision.
Let’s explore. What can it mean to know something?
Imagine you see a person across the street and say “hey, I know that guy.” What do you mean by that utterance? Surely you must mean something like “I believe that person across the street is someone of whom I was previously aware.” [Except that you are the kind of person who would be entirely comfortable ending your sentences with a preposition.] It would be infelicitous to say “I know that guy” if you had no prior knowledge of that person. And if you said “hey, I know that guy” and someone responded “so this is the first time you’re seeing him?”, you would think that your interlocutor hadn’t heard you, was not sufficiently familiar with English or was just nuts.
You’re not born knowing everybody. So how do you get from a state of not knowing that guy across the street to knowing him? It seems to me that you must come in visual contact with that guy (in person or via a photograph) and when you subsequently see him across the street, you match the stimulus (i.e., the visual appearance of the guy across the street) to your mental database and find that the stimulus sufficiently matches at least one entrant in your database. In other words, you know him now because you knew him before. [Note: Obviously sufficiency is not binary. There are various degrees of certitude. You might find that a person looks very familiar but something about their appearance makes you unsure – say a new haircut. The closer the match between the entry in your mental database and the live stimulus, the more certain you will be.]
Of course, there are multiple ways to gather your initial data and then to be exposed to the stimulus. Imagine a friend told you about some guy named Joe who had a wife named Josephine, three kids, a dog named Butch and worked as a beekeeper in St. Paul. Then you’re sitting in a bar in Shreveport and you overhear a conversation. A guy is talking to some woman and clearly trying to pick her up. You overhear him talking about his three kids, his dog named Butch and his rewarding career as a beekeeper in St. Paul. You might get up and say “Joe, don’t you think Josephine would be pissed if she knew you were trying to pick up women in Shreveport?!”
How could you do this? No, I don’t mean how could you stand between Joe and a night of pleasure. Of course you could do that. You, my dear reader, have scruples. I mean, how could you come to the conclusion that this guy you’ve never met or seen is the Joe of whom your friend spoke? Clearly, you must be matching the current stimulus to your database and finding a pattern match of sufficient strength. Again, you know this is Joe now because you knew him before.
Justice Potter Stewart’s famous non-definition of pornography is another example. While the judge clearly was having a rough time with it, his assertion simply cannot be accepted. I’m sure the judge was a very smart man. What he must have meant was that it was extremely difficult to provide a sound explicit definition of the term (particularly since defining this term was not merely an act of philosophy but one of politics) but that he had an implicit definition. In other words, he had a mental representation of it that he was unable to explicitly articulate. Otherwise, how can we possibly explain his ability to look at a plethora of stimuli and decide which were and which were not pornography?
What about color? Imagine someone shows you a swatch of color and you identify it as “blue.” How do you do that? How do you know it is blue? Does it make sense to assert that you can make such an identification without having any prior knowledge of blue or blueness? Either you’ve seen objects of similar chromaticity before and heard the word “blue” associated with such objects, or you have learned that “blue light” has a wavelength of about 475 nanometers and you’ve got a spectrophotometer to use on the swatch you’ve now been presented with. Again, you know it now because you knew it before.
Does anyone really believe that children learn to identify colors by seeing them and suddenly finding a name for them that happens to correspond to the name that their culture already uses?! Does anyone think that children go from a state in which they have absolutely no knowledge whatsoever of chairs to a state where they know what a chair is, by simply seeing one? Clearly not. Clearly there is knowledge of chairs or chairness in the mind of the child prior to their first act of identification of a specific chair. There must be some implicit definition. Plato would have called this a prior knowledge of forms. I will leave to others the question of whether Plato provides the best epistemological account. But I think it undeniable that we come to an act of perception with pre-existing knowledge – without which, perception would be quite impoverished. Modern cognitive research offers powerful evidence that the tabula rasa of old cannot possibly account for the demonstrable cognitive skill of human infants. Again, we bring quite a lot to the party. Even if we are unaware of our own knowledge.
So if we don’t believe that children suddenly know a chair when they see one without ever having any knowledge of what a chair is, then why would we believe that adults can know something the moment they see it without having any prior knowledge of the thing? It seems clear to me that we already know it. The act of definition is simply rendering this implicit sense into an explicit articulated definition. Definition transforms an implicit set of mental criteria into an explicit social object.
The only alternate explanation for all of these cases that makes sense to me is that the assignment to some category (that guy Joe, blue, pornography, etc.) is simply random. I suppose that is possible. But it is not knowledge. “I know it when I see it” would not be a fair description of a random act of categorization.
One last thought: I mentioned above that “I see it when I know it” would be more apt – both for cognition and vision. There is primitive sensation that occurs in our body and brain. But perception is a higher order phenomenon that is heavily dependent on prior knowledge. We don’t just see amorphous shapes. We see objects that we recognize. I see my father. I see a boat. I see an apple. Our ability to complete an act of perception depends on prior knowledge. There are many cool examples of our prior knowledge subverting perception in a way. These are called illusions – visual or auditory. These illusions only prove the point that prior knowledge plays a critical role in perception. They work because our brain interprets sensory data to conform with prior knowledge – even in some cases where such an interpretation is false.
The human mind is amazing. We know so much more than we are aware of. Defining a term can be mentally exhausting. But when getting it right matters, I feel quite strongly that definition is worth it. You don’t know it when you see it. You know it only when you know it.
{ 3 comments… read them below or add one }
You make a great case, an unarguable one, given the premise you set up. But I challenge the premise. I think there is something between, “I know nothing and wont define it/I’ll know it when I see it” and “I will define it and there for know it”. Some things can’t be defined to an absolute, so the best we can, and should, do is attach some meaning and principles to it and the rest will become clear in the experience of it. Stated differently, I think the real debate is in the definition of “definition”.
Lets take blue. My favorite color is blue. You know that sky blue, but not the really light one. The one that’s a little dense, but bright. Not bright like electric, bright, like it has a pinkish undertone. A pink undertone that you may see as a purple undertone, undertones are funny that way. Maybe better to describe it as periwinkle, but still more blue than grayish purple. OK, got it? I have defined my favorite blue. Now go to Home Depot and buy me a gallon of paint to match that color. Chances you would come home with the wrong one. Your filters preconceived notions and perceptions are a bit different than mine. But it would be entirely reasonable for me, having set up a number of descriptors and parameters to say, “that’s the best I can do in defining what I mean by my favorite blue, I will go to the store and get it myself, because I’ll know it when I see it.” The point being, and I think you would agree, some things just cant be defined in a way that insures everyone will see it the same way. This is the inherent nature of things that are subjective vs objective. Concepts are subjective. Of course we can, and should, in certain circumstances put criteria, parameters, markers, principles and guidelines around them, but to think we will ever get a definition that neutralizes the differences that exists in the “eyes of the beholders” is as absurd as thinking that if you have no prior knowledge of something and NO preconceived notions, you will “know it when you see it”.
Any concept that has attached to it any sense of personalization and relativity is subject to interpretation that can not be neutralized by a clear and definitive definition.
Somewhere between, “things must be definitely defined” and “some things can not be at all”, lies the likely truth, which is that all things can be defined to some extent, the amount of which depends on what you are trying to define (objective subjective) and to what end are you trying to define them (for personal reference or aligning a group).
I do believe that gut and intuition play a role in identifying and knowing something when you see it, but equally believe that even gut and intuition are based on some set of experiences from the past (previous lives included- ha ha). So to have NO prior knowledge and think you’ll know it when you see it is absurd, but one must acknowledge that abstract concepts can not be defined to an absolute, and the attempt to do so is equally absurd. We have a choice as to what we want to define and to what extent. That is not mental laziness, that is a gauge as to how much definition is really required in a given circumstance. As a practical matter, attempting to definitively define things that will always have some level of subjectivity will inevitably result in diminishing returns. How much is the right definition for an abstract concept, hmmm, not quite sure, but I’ll know it when I see it.
Josette:
Thanks for the very thoughtful comment.
I agree with you that there is something between, as you say, “I know nothing and won’t define it/I’ll know it when I see it” and “I will define it and therefore know it.” What lies between them is, as I pointed out in my post, an implicit definition. A mental representation of a concept.
Let’s go back to color. You offer an example of trying to describe a particular shade of blue. You point out, correctly I think, that if you were to attempt to describe this shade of blue and I went off to buy the paint, that I would likely come home with the wrong color. Where I think you get it wrong, is in using this example as a proof that blue (or this shade of blue) cannot be defined. You further argue that some things are subjective and I assume you would include your shade of blue (and therefore all others) in the category of subjective. Nothing could be further from the truth. Blue is completely objective. Not the name, that is merely a societal convention. But the thing itself can be measured and described in entirely objective terms. Color can be described in terms of hue, saturation and luminance. Or you could provide a Pantone number, etc.
The problem we would have in your blue paint example is not that your dream shade of blue cannot be defined. It can. With complete precision. What is lacking is not a definition, but the language in which to express it. The problem lies in trying to use a language that is less precise than the phenomenon it is intended to describe. If you aren’t that picky about your shade of blue then you can just use the word “blue” along with one of the layperson modifiers such as “light” or “dark” or “navy”. If you are picky then you need a language that is robust enough to convey the nuance of what is in your head. Much like the old story about Eskimos having 20 different words for snow, designers have such a robust language for color and I have provided some examples above.
But let’s come back to what is in your head. Surely we agree that if you do not use an appropriately robust language to define your shade of blue and I go to Home Depot for you, that I will return with the wrong color. But what if you go? Will you get the right color (assume they have it)? If you say yes, then I argue that you have clearly defined that color. And it is defined in some mental “language” – it is a mental representation of some phenomenon in the physical world. Our inability to communicate about that shade of blue is not a problem of definition but a problem of language. If you tell me that you might not be able to get the right color, then I would argue that you really haven’t selected a particular blue in your mind or that you did (and had it defined mentally) but then forgot.
I will move on to another point. You suggest we can never arrive at a definition for concepts that will neutralize differences in the eyes of beholders. I agree. But that simply means that people have – and will continue to have – differing definitions. It does not mean they have no definitions. Let’s take the concept “fair system of taxation”. Clearly this is a controversial topic in the US. Some people seem to define it as “a system that secures the outcome of an equitable allocation of resources” and others as “a system that collects from each taxpayer an equal share of their income”. And there are many other definitions. We will likely never reach consensus on this issue. But that does not mean each of us cannot define what we mean and then discuss with others on the basis of clarity.
“I’ll know it when I see it” when used to suggest that there can be no definition is a copout. And it is unfair. Take Potter Stewart’s pornography. Imagine you work in a profession that plays around in the boundaries or front lines of the pornography issue. Painter, radio host, writer, comedian, etc. And imagine you genuinely want to stay within the bounds of the law. Do you really want to be at Potter Stewart’s whim? Do you want to go about your job never knowing whether what you create today will land you in jail or not? Do you not have the right to ask for a clear definition?
This is an extreme case but all other cases are similar. If you refuse to define your terms (for any reason) then how can I possibly know what the heck you mean? Let’s go back to your favorite blue. Imagine you aren’t asking me to buy the paint for you but are simply talking about how your new home will look. You have one image in your head and I another. Now imagine that same mental gap for every word you use in every conversation. How can conversation ever be meaningful without definition? Can it even really be conversation?! It is not my objective for us all to reach consensus on every issue, every definition. It is my objective to foster intelligent conversation and debate. This can never happen so long as one side refuses to define its terms. To be clear, not all conversations need to be intelligent. Some are merely social exchange. Requiring explicit definition for every term ever time would mean we wouldn’t get very far. But sometimes precision counts. Sometimes concepts are critically important. On those occasions, I argue that people ought to do a much better job of making their implicit definitions more explicit.
So we are in violent agreement. I agree that not being willing to discuss individual definitions is a copout. I also agree that when trying to rally a group, an organization or even a friend in a particular direction, there needs to be a set of common agreements attained through intelligent conversation.
Im just leaving space for the reality that even with our best efforts, individual biases will leave gaps. Sometimes those gaps are OK, as the thing being known or seen is for the sole use of the person seeing or thinking they know it…But when it comes to organizational alignments, the resolution of that gap can be “we’ll know it if we see it”, at which point, by the way, another intelligent conversation is in order to determine if what we think we know when we see it, is in fact what we know it to be…and that we are seeing it the same way. And it is worth discussing till there is agreement or we risk, as an organization, diverging on critical points we ought to be converging on.
As an afterthought on color, I agree if I am an engineer or a scientist or a graphic designer I can define my favorite blue in terms of a professional…light wave frequencies, pantone #s and the like. But as a layman, I still hold that my blue is my blue, and only I will know it when I see it….and yes, it is because I have a clear picture in my mind of what it is. And I do not have the language available to me (without my laptop of course) to describe it or define it empirically….nor do I feel the need to, unless of course we are picking a color for our home and we need to both be happy with it, though we don’t both need to see it the same way…
Thanks as usual for provoking interesting and original thinking…reading your stuff and discussing it with you makes me feel smarter. Its like a great work out for my brain. Now I need to find that same mojo and apply it to working out for my body 🙂
{ 1 trackback }